Erv Wilson

[jancivil]

I am asking you to show a good reason for a preference for the piano's notes; I don't know of one... exception = that the ET fifth is close enough for rock and roll, but it's there with open strings on a fiddle anyway, and it's probably better than ET. NB: I can detect by ear deviations finer than one cent.

I'm saying chords are intoned pure or close to pure but the intervals between non-simultaneous notes cannot be pure.

Well, what you said implied that there is a preference for ET as measured vs just as measured. I know for sure no one measures either. It's a moot point. I asked you to show why, and twice you have only offered this same point, which has no real meaning.

I'm pretty sure you can't detect that fine deviations in melodic intervals.

That's so overconfident, it's just It's arrogant really.

In practice: A leading tone type of effect tends to be intoned sharper. A descending semitone tends to be intoned flatter. That's not theory, that's what happens. 12 tone ET is out the window.

Yes, that happens but it doesn't really show that the average places of the 12 notes in enharmonically equivalent music will significantly depart from ET. But I don't really know, you don't really know, I don't really care.

I know what I know from experience. Obviously no one measures these things in the heat of battle; I know enough of two traditions to know, eg., in Hindustani music they do measure intervals in pedagogy ("it's right here on the fingerboard, just so") and they care because it matters. Then there are people teaching violin in western concert music that in their pedagogy make a case of the enharmonics and their intonation, in harmony, and in melody owing some to the harmonic context. I know that I don't, in my own playing necessarily make a distinction between harmonic and melodic intervals and the theory I cited would apply (should I find a way to do so in virtual instruments) to either based in my knowledge.

That you don't has no bearing on any of that.

[jancivil]

I just re-rendered a piece where in one spot the piano was out of tune to a degree it was going to make me wince everytime I heard it. It was exposed so I heard the problem in terms of the melodic interval; in the solo editor I could hear it every time I hit the note before and that note. In concert with other instruments I will grant that the problem was exacerbated. I addressed it finally. I raised the pitch by '24' in the pitch bend lane. that is 24 steps out of 8192, which is the full range of pitch bend; in this case the engine's maximum is a whole tone, either direction. I got the calculator out; that's roughly .58 cents. Bugged the shit out of me. That was closer, and I found that lowering the velocity, maybe a couple layers got it to where I'm not going to cringe, and unless I focus on it am not displeased. Often I cut my losses...

In my electric guitar playing over the years, which is what informs what I do in the DAW, I have noticed all the time problems of intonation that derive from ET, and from the different phenomena owing to the construction of the instrument, the nut and so on, and how the string behaves here as opposed to there as it's a fretted instrument. I compensate by various means, and my tuning is a compromise. There is no perfectly set up guitar IME. I play lines, I'm not particularly concerned with chords. I don't make a distinction however, the note is true or it isn't. The correspondence, 'melody/harmony' does not vanish by talking about it.

I would recommend eschewing making assumptions of what others do based in what you understand.

[kalleaho]

I'm saying chords are intoned pure or close to pure but the intervals between non-simultaneous notes cannot be pure.

Well, what you said implied that there is a preference for ET as measured vs just as measured. I know for sure no one measures either. It's a moot point. I asked you to show why, and twice you have only offered this same point, which has no real meaning.

The I-IV-ii-V-I example shows that melodic intervals must be tempered, if we want to avoid shifting the tonic by a syntonic comma. That's a whopping 21.5 cents!

And when I say "tempered" it doesn't necessarily mean equally tempered. It could center around something close to quarter-comma meantone, for example. But in much of romantic repertoire the diesis 128:125 is also required to vanish which results in a center tuning close to 12-tone equal temperament.

I'm pretty sure you can't detect that fine deviations in melodic intervals.

That's so overconfident, it's just It's arrogant really.

OK, fine, you can play alpha-musicologist for all I care! Are you saying that you can detect that, for example, an ascending fifth is flat or sharp by half a cent (or even less)? That's just superhuman. There's no way anyone could (or wanted to) play that accurately, so real music must be really unbearable to you.